[Feature Request] Port Qwen2.5-VL Model

[jaytiwarihub]

I would like to work on porting the Qwen2.5-VL architecture. Since we already have the Qwen2 text backbone, I plan to implement the Vision Encoder and the C-Abstractor projector, then integrate them with the existing text backbone.

[laxmareddyp]

hi @jaytiwarihub,

Thanks for opening this issue! You’re welcome to start working on the model. I’ve assigned it to you.